Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Commonsense Reasoning on HellaSwag (Accuracy, Exit Position)
Loading...
59.1
Accuracy
Backbone
43.708
47.704
51.7
55.696
Apr 20, 2026
Accuracy
Exit Position
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Exit Position
Backbone
Backbone Model=Llama3....
2026.04
59.1
-
River-LLM
Backbone Model=Llama3....
2026.04
58.6
8.8
River-LLM
Backbone Model=Llama3....
2026.04
58.5
3.09
Ministral3 8B
Model Architecture=Min...
2026.04
58.5
-
River-LLM
Model Architecture=Min...
2026.04
58.2
2
River-LLM
Model Architecture=Min...
2026.04
58.2
2.38
Phi4-mini
Model Architecture=Phi...
2026.04
54.4
-
River-LLM
Model Architecture=Phi...
2026.04
53.9
4.14
River-LLM
Model Architecture=Phi...
2026.04
53.7
2.05
Backbone
Backbone Model=Llama3....
2026.04
45.1
-
River-LLM
Backbone Model=Llama3....
2026.04
44.9
10.94
River-LLM
Backbone Model=Llama3....
2026.04
44.3
3.71
Feedback
Search any
task
Search any
task